The Amiigo is a fitness band with a twist.

It promises, not just to track your activity, but to track your activities.

Spotting patterns in your movement data, the Amiigo team claims that they can distinguish jumping jacks from pull-ups.

They can even spot the subtle differences within a particular movement--like two types of bicep curls.

If the Indiegogo campaigners really can pull this off, it’d be hugely disruptive in the space.

And imagine receiving a warning when your form was getting sloppy in a way that could lead to injury.

If you’re willing to be an early adopter, Amiigo is looking for a few thousand users to help polish the device, after which time they’ll go after another wave of financing.

Co.Design

Kickstarting: A Fuelband That Could Name Your Exercises

A new startup attempts to crack the code on smarter fitness sensors that can pinpoint your exact exercise, automatically.

Everyone who owns a Nike+ Fuelband knows a leisurely walk to the store can score you more fuel than breaking a sweat on an elliptical. And heaven forbid you ride a bike; that’s completely worthless. Fitness bands are great at tracking running and walking, but aside from that, they’re basically clueless.

Amiigo is an Indigogo project that’s been in the works since early 2012. It’s a pair of accelerometers--one for your wrist, one for your shoe--that, like every other fitness device, track your movements and syncs them with a smartphone app. But what makes Amiigo different is an almost unbelievable amount of accuracy. The company claims that Amiigo won’t only distinguish walking, running, biking, or an elliptical; it can discern five different methods of bicep curls.

“Our primary competitive advantage lies in our ability to convert that raw accelerometer data into descriptions of specific activity,” cofounder Abe Carter tells Co Design. “You name it, we can track it.”

On paper, it all makes sense. To an accelerometer, each exercise appears as a specific pattern. Amiigo sends some engineers in to dig for this pattern inside absurd amounts of noise. A library of these reference patterns is placed on your smartphone. Then, when an Amiigo user tries an exercise for the first time, the software spots the closest reference pattern--actually swapping it out for the user’s own particular movement type. The software starts smart, then it customizes itself for optimal fidelity. You can even add your own exercises if they’re not on Amiigo’s list.

“One of the challenges is that once a pattern has been recognized, it’s very easy to pick out a chunk in the center of that pattern and analyze it carefully,” Carter explains. “What’s difficult is to figure out where that pattern initially started, and the last millisecond that it ended.”

Carter explains the challenge as a “plus or minus two reps problem,” creating software that’s smart enough not just to spot particular movement types but to do so instantly amongst huge amounts of white noise. This is where the company believes 1,000 Indiegogo users could come in handy.

Amiigo has incredible implications for fitness UX. Imagine a list of every exercise you completed at the gym. Consider sensors so accurate that they could warn you, based upon data of other users, that you were moving/lifting in a way that could cause injury. Or just think of how amazing it would be to graph all of this information to social media, so friends could see that you are “biking” or “on the elliptical.”

There’s just one catch--how could Amiigo possibly be pulling off this stunt when Nike, Jawbone, Withings, and FitBit are not?

“I’ve honestly thought that myself on a number of different occasions,” Carter admits. “I saw the announcement of the Fuelband and my heart sank. I didn’t know about the Fuelband before. I just don’t know why Nike hasn’t done this yet.”

Carter argues that there’s far too much focus on industrial design in this space, while there’s not enough emphasis on innovation behind the data itself (which can unlock the new user experiences we want). With all respect to the Amiigo, that’s an obvious perspective. The band itself is fairly horrid, maybe something you’d tolerate at the gym but certainly nothing you could wear with a straight face day in and day out.

Akram’s razor points to a simpler explanation, of course. Companies like Nike may be capable of identifying specific exercises, but they’re merely choosing not to. No doubt, requiring a second sensor to capture upper and lower body movements may be too much for the market to bear (in price and in juggling gadgets), so it could be an innovation being delayed to a day when every Nike shoe includes such a hardware standard. It’s just as possible that no company is reaching that accuracy tipping point--that 99/100 sweet spot--where the technology could be touted as a feature.

“It’s really not too good to be true,” Carter assures us. “[But] we’re happy that these bigger companies are a little bit slower. And we lucked out. Our guy doing a lot of the engineering on this is incredibly brilliant.”

We’ll try to test the Amiigo in the coming month. In the meantime, if you’re interested in supporting the project, the campaign is live now.

Order one here.

Add New Comment

1 Comments